INTRODUCTION - Mammalian (& non-mammalian) L1 (LINE-1) elements replicate (retrotranspose) by copying their RNA transcripts into DNA which is then integrated into the genome. The ~6 kb human L1 element has four regions: The 5?untranslated region (UTR) has a regulatory function; open reading frame (ORF) 1 encodes an RNA binding protein that can form multimers; ORF2 encodes a DNA endonuclease and reverse transcriptase; the 3?UTR contains a conserved G-rich polypurine motif. L1 replication produces mostly (~2/3) defective L1 copies, usually 5? truncated, that are retained in the genome and evolve as pseudogenes. L1 can also retropose the RNA transcripts of short interspersed repeated DNA (SINEs), & the processed transcripts from nuclear genes. L1 activity has persisted in mammals since before their radiation 100 million years ago (MYA) and has generated upwards of 40% of the mass of the studied rodent & primate genomes. In addition to the genomic impact of their considerable total mass, L1, and their SINE offspring, can affect the genome in other ways: They can participate in ectopic (non-allelic) recombination thereby causing genetic rearrangements, and upon insertion can inactivate genes or alter their regulation. Although L1 activity clearly has had a profound effect on the structure and function of modern genomes, little is known about the regulation of L1 activity, to what extent L1 and its host interact, or even whether L1 activity affects the fitness of its host. RECENT FINDINGS: THE AMPLIFICATION OF THE HUMAN SPECIFIC LINE-1 (L1) TA1 RETROTRANSPOSON HAS REDUCED THE FITNESS OF HUMANS - We earlier found that a human-specific L1 family (called L1Pa1 or Ta) emerged in humans soon after they and their closest ape relatives, chimpanzees, diverged about 6 Mya. Since then Ta continued to amplify and evolve & ~1Mya gave rise to the Ta1 subfamily. Essentially all of the L1 activity over the last 1 million years of the human lineage was due to the Ta1 subfamily. Although Ta1 insertions can cause genetic defects in current humans, these are rare, and it is not known whether Ta1 activity in general has been sufficiently deleterious to reduce the fitness of humans. This issue is of profound importance for understanding the effect of an L1 amplification on its host, & obviously for the long-term consequences of the Ta1 amplification on the well being of humans. Natural selection against a segregating allele provides near irrefutable evidence that it is deleterious. Therefore, in collaboration with Dr. Dimitri Petrov, we examined the unbiased collection of Ta1-containing alleles that we recently collected from four different ethnic populations for evidence of negative selection. We found that full-length (FL) Ta1 elements, but not truncated (TR) Ta1 elements or SINE (Alu) insertions, were subject to negative selection. As only FL L1 elements are capable of retrotransposition, L1 activity per se (or an effect of L1 uniquely related to its length) constituted a genetic burden for humans. We also found that negative selection increased as the Ta1 amplification has proceeded. Therefore, as has been theorized & found in model organisms like Drosophila, reduction in fitness increases after transposable elements reach some threshold number. We do not know how reduced fitness is exhibited in humans. However, as the Ta1 amplification is still ongoing in humans, we have no reason to conclude that the fitness of the human population is still not being reduced. L1 ELEMENT EVOLUTION IS DRAMATICALLY DIFFERENT IN MAMMALS & NON-MAMMALS: IMPLICATIONS FOR THE EFFECT OF L1 ON MAMMALIAN GENOMES - We had earlier shown that rat L1 families, as had others found for mice, generally evolve as a single lineage, a most unusual evolutionary scenario. We had also reported that most of the 16 primate-specific L1 families present in the human genome consist of a single lineage. Thus, each newly evolved L1 family supplanted its predecessor for replicative dominance & in modern humans only one, the Ta L1 family, remains active. We previously reported that this unusual pattern of L1 evolution was unique to mammals. Thus when we compared the evolution of mammalian & zebrafish L1 elements (in collaboration with Dr. David Duvernell) we found that, in dramatic contrast to mammals, L1 evolution in zebrafish produced at least 30 distinct L1 lineages (i.e., families), and, that although these families have remained active for more than 400 Myr, they generated only about 5% of fish DNA. Following up on these previously published results, we recently compared the mass of DNA generated by L1 retrotransposons in various mammalian & non-mammalian genomes using both published data from whole genome studies and unpublished analysis kindly provided by Arian Smit (personal communication). These results showed that the discrepancy between mammals & non-mammals held for every species (e.g., vertebrates, invertebrates, plants) studied. This further supports our earlier suggestion that, for some reason mammals are unique in their ability to tolerate large amounts of interspersed repeated DNA sequences, and that this permissive environment permitted the emergence of L1 elements that are active enough to compete for limited host resources. This competition could explain not only why mammalian L1 is generally limited to a single lineage but also why non-L1 retrotransposable elements have been driven largely to extinction in mammals. INTERACTION BETWEEN L1 AND ITS HOST - The evolutionary results presented above suggested that L1 might interact with host factors. Also, we had earlier found that the coiled coil motif of the L1 ORF1 protein (ORF1p) had undergone episodes of adaptive evolution early in hominid evolution. As adaptive evolution often implies an interacting system (e.g., a virus & its host), and as coiled coil domains can mediate protein-protein interaction, evolutionary change in this motif could reflect interaction of ORF1p with host factors. Using ORF1p as bait in a yeast two-hybrid assay, we identified eleven ORF1p-binding proteins from human cells. Of these, eight did not interact with the ancestral (i.e., pre-adapted) version of the ORF1p coiled-coil domain. Clearly then evolutionary change in the host has accompanied the evolution of L1. Six of the eight proteins that interact with ORF1p are thought to bind RNA, the primary substrate for L1 mediated retrotransposition. In other experiments we found that the coiled coil domain is absolutely essential for L1 retrotransposition and that amino acid changes in the coiled coil render it inactive for retrotransposition, but still support ORF1p interaction with itself & host factors. Recently we found that some ORF1p interactions also occur in a mammalian 2-hybrid system and are using this assay and others to further assess the biological significance of the interactions between ORF1p and host proteins. We have also progressed with our studies on the interaction between L1 RNA and the host protein, nuclear exchange factor (NXF) 1. NXF1 mediates nuclear export of non-spliced RNAs, as would be the case for L1 and retroviral RNAs. Preliminary results show that a region of the 3? end of ORF2 along with the contiguous 5? region of the 3? UTR, including the conserved G-rich polypurine motif, is required for this interaction. More detailed structural analysis for the requirements of this binding is now underway. As L1 elements in mammals totally dominate the retrotransposon landscape, this could result from L1 pre-empting NXF1.